ABSTRACT

In this paper, a simulator based on the Matlab is proposed to simulate the manipulation action of UVMS(Underwater Vehicle-Manipulator Systems).In the simulator a fuzzy adaptive PID controller and a relative position observer are presented. The observer can estimate the relative position between the vehicle and the target to be manipulated by fusing vision and inertial measurements. Simulation results indicate that the compounded control strategy can complete the manipulation task and manipulation error can reach the centimeter level.

INTRODUCTION

UVMS(Underwater Vehicle-Manipulator Systems) is an underwater vehicle system mounted one or more manipulators to complete certain underwater task. With the increased interest in the development of undersea intervention technology in the fields of offshore oil, military, and ocean scientific investigation, many approaches have been presented to improve undersea intervention capability of UVMS. Professor Antonelli investigated many aspects about UVMS, which included inverse kinematics resolution to find suitable vehicle/joints trajectory corresponding to a desired end effector (Antonelli and Chiaverini, 1998), adaptive tracking control(Antonelli and Caccavale, 2004) and interaction control of UVMS(Antonelli,2003). Mclain(1995) developed the hydrodynamic model of underwater manipulator in Standford. Professor J.Yhu(1998) introduced the development of SAUVIM(semi-Autonomous Underwater Vehicle for Intervention Mission)in the University of Hawaii. In the literature about UVMS, few papers have addressed a integrated simulator which includes the inverse kinematics and dynamics control as a whole.

However, this work can be very helpful in following aspects:

  1. tesing multiple motion trajectories under dynamic control and selecting the best one,

  2. selecting the sensors to be used in UVMS and analyzing the influence of sensor's performance to UVMS's manipulating task,

  3. providing great insight into the control method because the parameters of modeling can be varied easily and all measurements are immediately available.

This paper mainly addresses a simulator whose main issues are localization and manipulation.

Keywords:
artificial intelligence, target point, body frame, adaptive pid controller, underwater vehicle-manipulator system, observer, uvm, vehicle, inertial measurement, controller
Subjects:
Reservoir Simulation, Information Management and Systems, Dynamic model update algorithms, Artificial intelligence
This content is only available via PDF.
2008. The International Society of Offshore and Polar Engineers
You can access this article if you purchase or spend a download.